My last 'non Amd 3' CU Install.

[Dave OCD]

Put this in today, I hope it doesn't burst into flames.:tongue3: Didn't have the time to faff about and get the cabling 'perfect' as I had so much to do today but I think it's ok. With the BG boards I generally 're number' the earth bar as to me it's illogical having the CPCs running in the opposite order of the neutrals,I'll miss the insulated CUs. Sorry about the
poor photo.

 

[shanky887614]

So this rcd could be leaving the install in a dangerous state untill the next test, and the next test always gets done on due date lol. I stick to my guns a 100mA type S upfront on a domestic TT at origin is the best way to install that system

how? it is relatively easy in a lot of places to get a really low ra so why is an additional rcd needed in all cases

 

[Simonslimline]

What a shocking end product.? If i left that sort of end product with my customers id have none left, funny how standards differ from one end of Country to other!!

Any chance of uploading a photo/s of the standard of your work?

Maybe we can all learn something.

:sifone:

 

[telectrix]

nobody's disputing that it's the best way, just not required by regs.

 

[hightower]

So this rcd could be leaving the install in a dangerous state untill the next test, and the next test always gets done on due date lol. I stick to my guns a 100mA type S upfront on a domestic TT at origin is the best way to install that system

Why type s? I thought 100ma would offer enough discrimination from a 30ma? Read somewhere 3 times as much, so for discrimination 90ma or higher should act independently without the need for a time delay

 

[Midwest]

So this rcd could be leaving the install in a dangerous state untill the next test, and the next test always gets done on due date lol. I stick to my guns a 100mA type S upfront on a domestic TT at origin is the best way to install that system

As Tel says, believe this was a common design a few years ago and perhaps a few sparks still use it. You may find that difficult with A3? The same could be said of the reliability of MCB's, at least an RCD can be tested. What if the S type goes faulty, how do you design that out

 

[Simonslimline]

Why type s? I thought 100ma would offer enough discrimination from a 30ma? Read somewhere 3 times as much, so for discrimination 90ma or higher should act independently without the need for a time delay

Where did you hear that mate?

Should be S type.

 

[Midwest]

Why type s? I thought 100ma would offer enough discrimination from a 30ma? Read somewhere 3 times as much, so for discrimination 90ma or higher should act independently without the need for a time delay

Can't give you the science others will, but time delay rcd is a better design, as I've been advised before.

 

[hightower]

Where did you hear that mate?

Should be S type.

Will have a flick through my books and see if I can dig it out

 

[shanky887614]

hightower i recommend you have a look on google at discrimination. if you have a 100ma ac rcd and then a 30ma ac rcd they will both trip at once on fault.


an s type or time delayed is recommended because it allows any rcd's/mcbs down stream to clear the fault without tripping.

130-500ms for wylex s type rcd.

 

[davesparks]

Why type s? I thought 100ma would offer enough discrimination from a 30ma? Read somewhere 3 times as much, so for discrimination 90ma or higher should act independently without the need for a time delay

An S type is used to ensure discrimination in all cases as far as possible.
Without a time delay fault currents could easily cause both devices to operate simultaneously. Also any background earth leakage on individual circuits could add together to bias the 100mA to trip first without the time delay.

 

[Simonslimline]

Will have a flick through my books and see if I can dig it out

As an example - The first company i worked for had an office in a small industrial unit, TT installation with an upfront 100ma RCD, SWA sub-main from switch room to unit. A new consumer unit was put in the office we used and had 30ma RCD's. Both RCD's would operate in the event of a fault, Resulting in going to the switch room to reset.

 

[hightower]

Hmmmm, just gone to the page about selectivity in the book I'm currently reading and you guys are right - it says it should be s type in this case. In going to continue flicking through my books as in not sure what the hell I've read giving me this idea

 

[Leesparkykent]

Why type s? I thought 100ma would offer enough discrimination from a 30ma? Read somewhere 3 times as much, so for discrimination 90ma or higher should act independently without the need for a time delay

An S type RCD is also known as a time delayed RCD. If a 100mA RCD is installed upstream of a 30mA RCD, there will be no certainty as to which RCD will trip first. Cascading of General Type RCDs is therefore not recommended. Discrimination can only be provided by ensuring that the upstream RCD has a trip time and trip level envelope which is completely outside that of the downstream RCD. This can only be assured by using an S Type RCD upstream and General types downstream.

 

[hightower]

Last paragraph is where I'd read it, doesn't mention s type until much later hence my confusion

 

[shanky887614]

what book is that hightower?